Table 3 Multivariable logistic regression analysis of rebound in positive subscores on the abnormally low prolactin levels among schizophrenia patients participating in the trial of switching to aripiprazole (N = 63)
Variables | Rebound in positive subscores (n = 21) | No rebound in positive subscores (n = 42) | Multivariate-adjusted OR (95% CI) | |
|---|---|---|---|---|
Model 1 | Model 2 | |||
Male, n (%), (ref. female) | 9 (42.90) | 17 (40.50) | 0.57 (0.16–2.06) | 0.42 (0.09–2.01) |
Early age of onset, n (%), (ref. late age of onset)a | 2 (9.52) | 6 (14.29) | 0.62 (0.10–3.94) | 0.41 (0.06–2.82) |
Preswitching medication | ||||
First generation antipsychotics, n (%) | 12 (57.10) | 13 (31.00) | 1.00 (reference) | 1.00 (reference) |
Second generation antipsychotics, n (%) | 9 (42.90) | 29 (69.00) | 0.31 (0.10–0.98) | 0.22 (0.06–0.82) |
Prolactin level at baseline, ng/dL, mean (SD) | 47.6 (69.48) | 55.6 (57.63) | ˍ | 1.00 (0.99–1.01) |
Positive subscore in PANSS at baseline, mean (SD)b | 12.2 (4.41) | 9.4 (4.26) | ˍ | 1.25* (1.07–1.46) |
Abnormally low prolactin levels at follow up, n (%) | 12 (57.10) | 13 (31.00) | 3.55* (1.02–12.5) | 4.95* (1.08–22.7) |
Interaction term, abnormally low prolactin levels × sex | 0.66 (0.05–9.23) | 0.38 (0.02–7.24) | ||